ISSN print edition: 0366-6352 ISSN electronic edition: 1336-9075 Registr. No.: MK SR 9/7 Published monthly

Optimizing propylene production via 2-butene metathesis: catalytic efficiencies and AI-driven process enhancement

Amin Hedayati Moghaddam, Morteza Esfandyari, Hossein Sakhaeinia, and Abdellatif Mohammad Sadeq

Department of Chemical Engineering, Central Tehran Branch, Islamic Azad University, Tehran, Iran

E-mail: ami.hedayati_moghaddam@iau.ac.ir

Received: 2 September 2024  Accepted: 30 July 2025

Abstract:

This study focuses on the metathesis of 2-butene to convert low value products into higher value propylene. In this work, using machine learning (ML) techniques, several robust models were built and developed to predict the mole fraction of components in products of catalytic metathesis process over WO₃ on mesoporous support without the need for profound knowledge about exact reaction mechanism and their kinetics. The operative parameters were reaction temperature and residence time. The process performance was assessed using conversion and product selectivity as responses. Cross-validation technique was used during the model development. The developed models were used to investigate the mechanism of process as well as examining the effects of operative parameters on process performance. Further, these models were used to optimize the process in companion with genetic algorithm (GA).

Keywords: Metathesis; Propylene; Modeling; Artificial intelligence; Machine learning; 2-Butene; Genetic algorithm

Full paper is available at www.springerlink.com.

DOI: 10.1007/s11696-025-04282-3

Chemical Papers 79 (11) 7713–7724 (2025)